Innate lymphoid cells (ILCs) are highly plastic and predominantly mucosal tissue-resident cells that contribute to both homeostasis and inflammation depending on the microenvironment. The discovery of naïve-like ILCs suggests an ILC differentiation process that is akin to naïve T cell differentiation. Delineating the mechanisms that underlie ILC differentiation in tissues is crucial for understanding ILC biology in health and disease. Here, we showed that tonsillar ILCs expressing CD45RA lacked proliferative activity, indicative of cellular quiescence. CD62L distinguished two subsets of CD45RA + ILCs. CD45RA + CD62L + ILCs (CD62L + ILCs) resembled circulating naïve ILCs because they lacked the transcriptional, metabolic, epigenetic, and cytokine production signatures of differentiated ILCs. CD45RA + CD62L − ILCs (CD62L − ILCs) were epigenetically similar to CD62L + ILCs but showed a transcriptional, metabolic, and cytokine production signature that was more akin to differentiated ILCs. CD62L + and CD62L − ILCs contained uni- and multipotent precursors of ILC1s/NK cells and ILC3s. Differentiation of CD62L + and CD62L − ILCs led to metabolic reprogramming including up-regulation of genes associated with glycolysis, which was needed for their effector functions after differentiation. CD62L − ILCs with preferential differentiation capacity toward IL-22–producing ILC3s accumulated in the inflamed mucosa of patients with inflammatory bowel disease. These data suggested distinct differentiation potential of CD62L + and CD62L − ILCs between tissue microenvironments and identified that manipulation of these cells is a possible approach to restore tissue-immune homeostasis.

No abstract

Background The origin of human Innate Lymphoid Cells (ILCs) remains less well characterised compared to mice. Single-cell RNA-sequencing recently revealed a population of ILC precursors (ILCP) expressing CD62L and CD45RA, which shows diversity in frequencies among blood, tonsil and intestine. Objective We set out to determine the differentiation capacity and epigenetic landscape of blood, tonsil and intestinal ILCP. Methods Blood, tonsil and intestinal ILCP were sort-purified on the basis of CD62L and CD45RA (CD62L+/−CD45RA+/−) and cultured on OP9-DL1 stromal cells under specific cytokine conditions. Additionally, genome-wide chromatin accessibility was measured using ATAC-sequencing. Results We show that all ILCP subsets in blood and tonsil were able to differentiate to IFN-γ-producing ILC1/NK cells. Whereas all blood ILCP subsets readily differentiated to IL-13-producing ILC2, no or very few ILC2 were generated from tonsil ILCP. In contrast, tonsil ILCP subsets showed higher ability to generate IL-22-producing ILC3, compared to blood ILCP. These data indicate that ILCP in blood are poised towards ILC1/ILC2 while ILCP in tonsil favours ILC1/ILC3 differentiation. To further dissect their development, we are currently analysing global epigenetic information of tonsil ILCP, aiming to delineate their chromatin landscape. Finally, we investigated the ILCP in intestinal samples and detected accumulation of CD62L−CD45RA+ ILCP in the inflamed mucosa of patients with inflammatory bowel disease (IBD) as compared to the non-inflamed tissue. We are currently analysing the potency of those cells. Conclusion Heterogeneity in ILC differentiation in blood and tissues could further reveal novel targets for IBD treatment.